Anthropic began restructuring its leadership and organization in early 2026 as it prepares for a potential IPO. The company, which could go public as early as October at a valuation of up to $630 billion, has recently pulled ahead of competitors in revenue. Its annualized run rate (a prediction of annual revenue based on current quarterly performance) surpassed $30 billion in April, more than triple the figure at the end of 2025 and ahead of OpenAI’s reported $25 billion. Headcount has also surged, with roughly 2,300 employees at the end of last year, more than double its size just months earlier. Together, these changes reflect a growing emphasis on both rapid experimentation and commercial scale, while maintaining safety as a core differentiator.
A key part of this restructuring is the creation of Anthropic Labs in January, a research and development unit focused on incubating experimental products at the frontier of Claude’s capabilities. One of its first major decisions was to withhold public release of its most advanced model, Claude Mythos, due to its ability to identify and exploit software vulnerabilities. Instead, the company is pursuing a limited rollout, known as Project Glasswing, granting access to more than 50 organizations, including Microsoft and Nvidia, to help strengthen cyber defenses.
These structural changes are closely tied to a broader leadership reshuffle, including the promotion of former chief product officer Mike Krieger to co-lead Anthropic Labs.
“Krieger is someone who knows how to build products that stick,” Div Garg, founder and CEO of on-device superintelligence company AGI, Inc., told Observer. “At this stage, the constraint isn’t model quality. The constraint is turning capability into things people actually use and pay for, repeatedly, at scale.”
Mike Krieger and Ben Mann co-lead Anthropic Labs
Krieger, Anthropic’s former chief product officer and a co-founder of Instagram, transitioned to co-lead Anthropic Labs at its launch earlier this year. At Instagram, he served as chief technology officer, and later co-founded the news app Artifact, which he sold to Yahoo in 2024. A native of Brazil, Krieger is a Stanford University alumnus.
Ben Mann, an Anthropic co-founder, previously helped architect GPT-3 at OpenAI and worked as a software engineer at Google. Before moving to Labs, he served as Anthropic’s lead product engineer, focusing on A.I. alignment and harm mitigation. Mann graduated from Columbia University.
At Anthropic Labs, Krieger and Mann oversee a range of high-stakes initiatives, including the controlled rollout and governance of Claude Mythos, the company’s most advanced model. While Mythos can significantly strengthen cybersecurity by identifying and exploiting software vulnerabilities, it also poses risks if misused. To manage that tension, Anthropic has opted for a limited release under Project Glasswing.
With Labs’ creation and their appointment, Anthropic has “the right structure in place to support the most critical motions for our product organization—discovering experimental products at the frontier of Claude’s capabilities and scaling them responsibly,” Anthropic president Daniela Amodei wrote in a release.
Ami Vora replaces Mike Krieger as CPO
Following Krieger’s move to Anthropic Labs, Ami Vora has taken on the role of chief product officer. She joined the company in December 2025 as head of product and was quickly promoted. Vora previously spent 15 years at Meta, where she held leadership roles, including vice president of product at Facebook and vice president of product and design at WhatsApp. She began her career at Microsoft and remains on the board of cloud monitoring platform Datadog.
As CPO, Vora works closely with chief technology officer Rahul Patil to scale Claude beyond experimentation and expand Anthropic’s market presence.
Rahul Patil stays on as CTO, with an expanded role
Anthropic’s broader leadership bench remains deep, with all seven co-founders still at the company. Rahul Patil, who became CTO in October, succeeded Sam McCandlish, now chief architect. Patil previously served as CTO of Stripe and has led engineering teams at Microsoft, AWS and Oracle. Now working in close coordination with Vora, Patil is focused on bridging the gap between technical research and production-ready products.
As Anthropic moves closer to a potential IPO, that alignment is increasingly critical. “Anthropic has decided the frontier lab model only gets you so far,” said AGI, Inc.’s Garg. “My read is that Anthropic is preparing for a more competitive commercial phase, probably regardless of IPO timing.”
Other executives shaping Anthropic’s future
- Dario Amodei, CEO and co-founder: Dario Amodei previously served as vice president of research at OpenAI. He founded Anthropic in 2021 with his sister, Daniela, and other former OpenAI colleagues.
- Daniela Amodei, president and co-founder: Daniela Amodei, who previously served as vice president of safety and policy at OpenAI, oversees Anthropic’s core operations, including chief technology officer Rahul Patil and chief architect Sam McCandlish.
- Jared Kaplan, chief science officer and co-founder: Anthropic co-founder and former OpenAI researcher Jared Kaplan serves as chief science officer. Since 2024, he has also served as the company’s responsible scaling officer, helping guide safety-related decisions.
- Jan Leike, alignment science lead: Jan Leike, who co-led OpenAI’s superalignment team, has served as Anthropic’s alignment science lead since 2024.
- Sam McCandlish, chief architect and co-founder: Another former OpenAI employee, Sam McCandlish focuses on model training and large-scale systems development. He previously served as Anthropic’s CTO.
- Tom Brown, chief compute officer and co-founder: Former OpenAI GPT-3 researcher Tom Brown oversees Anthropic’s compute infrastructure.
- Vitaly Gudanets, CISO: Vitaly Gudanets has served as Anthropic’s chief information security officer since September. He previously led security efforts at Netflix.
- Jack Clark, head of policy and co-founder: A former OpenAI policy director and technology journalist, Jack Clark leads Anthropic’s policy work.
- Krishna Rao, CFO: Krishna Rao joined Anthropic as chief financial officer in 2024 after previously leading finance at Airbnb.
- Christopher Olah, interpretability research lead and co-founder: Christopher Olah, a former interpretability lead at OpenAI, heads Anthropic’s interpretability research, focusing on model transparency and A.I. safety.
Anthropic’s board and trust
In February, Anthropic appointed Chris Liddell, a former deputy White House chief of staff and former CFO at Microsoft and General Motors, to its board of directors. Daniela Amodei said Liddell has “a track record of helping organizations get [technology, public service and governance] right when the stakes are highest.”
The rest of the board remains unchanged, including Dario Amodei, Daniela Amodei, Yasmin Razavi, Jay Kreps, Reed Hastings and Chris Liddell. The Long-Term Benefit Trust recently removed Kanika Bahl and Zach Robinson and added Mariano-Florentino Cuéllar; Neil Buddy Shah remains on the trust board.
